Chinese herbal medicine intelligent warehousing system and method based on AI and block chain

Abstract

The invention discloses a Chinese herbal medicine intelligent warehousing system based on AI and blockchain, which comprises an Internet of Things sensing module, an AI processing module, a blockchain storage module and a user interaction module, and is characterized in that the Internet of Things sensing module is used for data acquisition, the AI processing module is used for data analysis and decision making, and the blockchain storage module is used for blockchain storage. The block chain storage module is used for ensuring the security and non-tampering of data, the user interaction module provides a friendly operation interface, and the system is characterized in that the system can automatically optimize the storage layout, realize whole-course quality tracing and dynamically adjust the inventory level.

Description

Technical Field

[0001] The present invention relates to the field of information management systems, and in particular to an intelligent storage system and method for Chinese herbal medicine based on AI and blockchain. Background Art

[0002] With the enhancement of global health awareness and the international dissemination of traditional Chinese medicine culture, the traditional Chinese medicine industry has ushered in unprecedented development opportunities. However, the traditional Chinese herbal medicine storage management model faces many challenges, such as low management efficiency, information islands, data opacity, and insufficient security, which seriously restrict the further development of the traditional Chinese medicine industry. In order to solve these problems, the present invention proposes a new intelligent storage system for Chinese herbal medicine based on AI (artificial intelligence) and blockchain technology, aiming to realize intelligent, efficient, transparent and safe storage management through technological innovation.

[0003] At present, the storage management of Chinese herbal medicine mainly relies on manual operation and simple information systems, and there are the following main problems: Low management efficiency: manual inventory, classification, picking and other processes are time-consuming, labor-intensive and prone to errors. Data islands: data between different links are difficult to share, forming information barriers. Information opacity: it is difficult for consumers to understand the source, processing process and quality information of Chinese herbal medicines. Insufficient security: data can be easily tampered with, storage environment monitoring is not in place, and there are quality risks.

[0004] The rapid development of AI and blockchain technology has made it possible to solve the above problems. AI technology is good at processing complex data and optimizing decision-making processes, while blockchain technology has unique advantages in data security and traceability due to its decentralized and tamper-proof characteristics. Combining the two and applying them to the storage management of Chinese herbal medicines is expected to achieve a comprehensive upgrade of storage management. Summary of the invention

[0005] The present invention provides an intelligent storage system and method for Chinese herbal medicine based on AI and blockchain. The intelligent storage system for Chinese herbal medicine consists of four major parts: an Internet of Things perception module, an AI processing module, a blockchain storage module, and a user interaction module. The modules work together to realize the intelligence and efficiency of storage management.

[0006] Furthermore, the IoT sensing module is responsible for collecting environmental parameters such as temperature, humidity, light, and gas concentration in the storage environment, as well as logistics information such as the entry, exit, and inventory status of Chinese medicinal materials. By deploying IoT devices such as sensors and RFID tags, real-time data collection and transmission can be achieved.

[0007] Furthermore, the AI ​​processing module receives data transmitted by the IoT sensing module and uses AI algorithms such as machine learning and deep learning to analyze and process the data. This module includes multiple sub-modules such as intelligent storage location recommendation algorithm, Chinese herbal medicine quality traceability model, inventory optimization management algorithm, etc., to provide decision support for warehouse management.

[0008] Furthermore, the blockchain storage module: constructs a distributed ledger based on blockchain technology to store and manage the full life cycle data of Chinese herbal medicines. Through hash functions, timestamps and other technical means, the data is ensured to be tamper-proof and traceable.

[0009] Furthermore, the user interaction module provides a user-friendly operation interface and data analysis reports for viewing and use by users such as warehouse managers and consumers. Users can query the source, processing process, quality information, etc. of Chinese herbal medicines through this module to improve information transparency.

[0010] Furthermore, the IoT perception module establishes a connection with the AI ​​processing module through a network communication protocol. The sensor and RFID reader send the collected data to the gateway or data concentrator in the form of a digital signal, and after preliminary processing and integration, transmit it to the server of the AI ​​processing module through the network. After receiving the data, the server of the AI ​​processing module stores it in the database and starts the corresponding AI algorithm for analysis. The AI ​​processing module is connected to the blockchain storage module through an API (application programming interface). The data analyzed and processed by the AI ​​processing module is encrypted and sent to the blockchain storage module through the API. After receiving the data, the blockchain storage module stores it according to the storage rules of the blockchain. At the same time, the blockchain storage module can also provide historical data for analysis and training according to the request of the AI ​​processing module to optimize the AI ​​algorithm. The user interaction module is connected to the blockchain storage module through the network. When the user initiates a data query request in the user interaction module, the request is sent to the blockchain storage module through the network. After verifying the identity and authority of the user, the blockchain storage module returns the corresponding data to the user interaction module. The user interaction module decrypts and parses the returned data and displays it to the user in an intuitive form.

[0011] On the other hand, the invention also provides an intelligent storage location recommendation method, which uses a greedy algorithm combined with a graph neural network (GNN) to optimize the location. Define the location selection function

[0012] F(i) = αD(i) + βT(i) + γ*C(i), where D(i) is the distance cost, T(i) is the time cost, C(i) is the capacity cost, and α, β, and γ are weight coefficients. By adjusting the weight coefficients, the optimization target can be flexibly set according to the actual situation.

[0013] The method specifically includes the following steps:

[0014] S1 collects real-time order data and inventory status information;

[0015] S2 uses GNN to predict the visit frequency of each cargo location in the future;

[0016] S3 gradually optimizes the cargo location allocation based on the cargo location selection function F(i) and the greedy strategy;

[0017] S4 outputs the optimal cargo location allocation plan to guide picking operations.

[0018] On the other hand, the invention also provides a method for tracing the quality of Chinese herbal medicines, which is based on the distributed ledger of blockchain and combines hash functions to build a traceability chain. Define hash function H(m) = hash(m), where m is the content of the data block. The data block content is converted into a hash value of fixed length through the hash function and stored on the blockchain.

[0019] The method specifically comprises the following steps:

[0020] S1 generates a unique hash value at each key link of Chinese medicinal materials such as planting, picking, processing, and warehousing; S2 records the hash value and corresponding metadata (such as time, location, operator, etc.) on the blockchain; S3 queries the hash value and its corresponding metadata on the chain when it is necessary to trace the quality information of Chinese medicinal materials; S4 builds a full life cycle traceability chain of Chinese medicinal materials based on metadata and displays it to consumers or managers.

[0021] On the other hand, the invention also provides an inventory optimization management algorithm: an inventory forecasting model that uses a moving average (MA) combined with a seasonal adjustment factor (S). The forecasting model is defined as Q(t) = MA(t)*S(t), where MA(t) is the moving average of the tth period, and S(t) is the seasonal adjustment factor of the tth period.

[0022] The method specifically includes the following steps:

[0023] S1 First, collect past sales data from the historical sales database, including key indicators such as sales volume and return volume for each month or quarter. These data will serve as the basis for algorithm training and prediction; S2 Moving average calculation: Use the moving average method (MA) to smooth historical sales data to reduce the impact of seasonal fluctuations and other random factors on the prediction results. Set a fixed time window (such as sales volume in the last three or six months), calculate the average sales volume in the window, and obtain the moving average MA(t); S3 Determination of seasonal adjustment factor: Determine the seasonal adjustment factor S(t) based on the seasonal variation pattern in historical sales data. This usually requires periodic analysis of historical data to identify the impact of different seasons (such as spring, summer, autumn, and winter) on sales, and calculate the adjustment factor for each season accordingly; S4 Inventory forecasting: Multiply the moving average MA(t) by the seasonal adjustment factor S(t) to obtain the inventory demand Q(t) for the forecast period. The forecast value will be used as the basis for formulating inventory strategies and replenishment plans; S5 Dynamic Adjustment: In actual applications, the system will update the moving average and seasonal adjustment factor in real time based on the latest sales data and inventory status to achieve dynamic adjustment of inventory forecasts. In addition, the forecast model can be manually intervened according to special events (such as promotions, epidemic outbreaks, etc.) to improve the accuracy of the forecast.

[0024] The present invention has the following beneficial effects:

[0025] (1) Deep integration of AI and blockchain: This invention innovatively combines AI and blockchain technology, using AI to process complex data and optimize the decision-making process, while using blockchain to ensure the security and immutability of data. This deep integration not only improves the level of intelligent warehouse management, but also enhances the reliability and credibility of data;

[0026] (2) Intelligent storage location recommendation: Through the combination of GNN and greedy algorithm, the optimal allocation of storage locations is achieved. The algorithm can dynamically adjust the storage location layout based on real-time order data and future forecast information to improve picking efficiency and inventory turnover. This intelligent storage location recommendation mechanism is of great significance to improving the overall efficiency of warehouse management;

[0027] (3) Full-process quality traceability system: A full-process quality traceability system from planting to warehousing has been established based on blockchain technology. This system can ensure that the source of Chinese medicinal materials is clear, the processing process is transparent, and the quality information is traceable. It has a positive effect on improving the market trust of Chinese medicinal materials and consumer satisfaction.

[0028] (4) Dynamic inventory optimization: Combined with historical sales data and seasonal changes, dynamic adjustment of inventory levels is achieved. This function helps reduce inventory backlogs and out-of-stock risks, lower inventory costs and improve capital efficiency;

[0029] (5) Unstructured data processing: AI image recognition and natural language processing technologies are introduced to effectively process and analyze unstructured data in the storage environment. These technologies can automatically identify and classify information such as the variety, grade, and quality status of Chinese medicinal materials, thereby improving the level of refined storage management;

[0030] (6) High degree of automation and unmanned operation: The integration of advanced technologies such as the Internet of Things, drones, and autonomous mobile robots has achieved high degree of automation and unmanned operation of warehousing operations. The application of these technologies reduces manual intervention and error rates, and improves the efficiency and safety of warehousing operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 System principle flow chart; DETAILED DESCRIPTION

[0032] The following embodiments of the present invention are described in further detail in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0033] Embodiment 1:

[0034] The specific architecture of the Chinese herbal medicine intelligent storage system based on AI and blockchain is as follows:

[0035] (I) IoT Perception Module

[0036] The IoT sensing module is the data collection front end of the entire intelligent warehousing system. It is like the "eyes" and "antennae" of the system, responsible for collecting environmental parameters such as temperature, humidity, light, and gas concentration in the storage environment, as well as logistics information such as the entry, exit, and inventory status of Chinese herbal medicines. In order to achieve comprehensive and accurate data collection, various types of sensors are deployed in the warehouse, such as temperature and humidity sensors, light sensors, gas sensors, etc. These sensors are distributed in different areas of the warehouse to monitor environmental conditions in real time. At the same time, each batch of Chinese herbal medicines entering the warehouse is equipped with RFID tags to record basic information such as its variety, origin, harvest time, and entry time. During the transportation and storage of Chinese herbal medicines, the logistics information in the RFID tags is automatically read and updated through the readers and writers installed at the warehouse entrances and shelves. The IoT sensing module transmits the collected data to the AI ​​processing module in real time through a wireless network or a wired network, providing a basis for subsequent data analysis and processing.

[0037] (II) AI Processing Module

[0038] The AI ​​processing module is the "brain" of the system. It receives data transmitted by the IoT sensing module and uses AI algorithms such as machine learning and deep learning for analysis and processing. Intelligent storage location recommendation algorithm: Based on the characteristics of Chinese medicinal materials (such as temperature and humidity requirements, volatility, etc.), the current warehouse location usage, and inventory turnover rate, the most suitable storage location is recommended for newly stored Chinese medicinal materials to improve warehouse space utilization and access efficiency. Chinese medicinal materials quality traceability model: Use deep learning algorithms to analyze the collected environmental data and logistics information, and establish a correlation model between the quality of Chinese medicinal materials and the storage environment and logistics process. When problems are found in the quality of Chinese medicinal materials, possible influencing factors can be quickly traced back to provide a basis for quality control. Inventory optimization management algorithm: By analyzing historical inventory data and sales data, the demand trend of Chinese medicinal materials is predicted, and a reasonable inventory strategy is formulated to avoid inventory backlogs or out-of-stock phenomena. The AI ​​processing module sends the results of the analysis and processing to the blockchain storage module and the user interaction module respectively to provide support for data storage and user decision-making.

[0039] (III) Blockchain storage module

[0040] The blockchain storage module is the data security and trust cornerstone of the system. It builds a distributed ledger based on blockchain technology to store and manage the full life cycle data of Chinese medicinal materials. When the IoT perception module and AI processing module generate new data, the data is first processed using a hash function to generate a unique hash value. Then, the data and its hash value are timestamped and packaged into a new block. New blocks are added to the blockchain through the consensus mechanism of the blockchain (such as proof of work, proof of equity, etc.). Due to the chain structure of the blockchain and the characteristics of the hash function, once the data is recorded on the blockchain, it cannot be tampered with. At the same time, the timestamp ensures the order and timeliness of the data, so that each piece of data has a clear time context.

[0041] (IV) User Interaction Module

[0042] The user interaction module is a bridge between the system and the user, providing a friendly operation interface and data analysis reports for warehouse managers, consumers and other users to view and use. Warehouse managers can view the warehouse's environmental parameters, inventory status, cargo distribution and other information in real time through computer or mobile applications, and perform operations such as warehousing, outbound, and inventory. Consumers can scan the QR code on the packaging of Chinese herbal medicines or visit related websites to query the source, processing process, quality information, etc. of Chinese herbal medicines to enhance their trust in the products. The user interaction module obtains the corresponding data from the blockchain storage module according to the user's authority and needs, and displays it to the user in the form of charts, reports, text, etc. At the same time, the user interaction module also receives the instructions and data entered by the user and feeds them back to other modules for processing.

[0043] The intelligent storage location recommendation method uses a greedy algorithm combined with a graph neural network (GNN) to optimize the location. Define the location selection function F(i) = αD(i) + βT(i) + γ*C(i), where D(i) is the distance cost, T(i) is the time cost, C(i) is the capacity cost, and α, β, and γ are weight coefficients. By adjusting the weight coefficients, the optimization target can be flexibly set according to the actual situation.

[0044] The method specifically includes the following steps:

[0045] S1 collects real-time order data and inventory status information;

[0046] S2 uses GNN to predict the visit frequency of each cargo location in the future;

[0047] S3 gradually optimizes the cargo location allocation based on the cargo location selection function F(i) and the greedy strategy;

[0048] S4 outputs the optimal cargo location allocation plan to guide picking operations.

[0049] The quality traceability method of Chinese herbal medicine is based on the distributed ledger of blockchain and combines hash functions to build a traceability chain. Define the hash function H(m) = hash(m), where m is the content of the data block. The data block content is converted into a hash value of fixed length through the hash function and stored on the blockchain.

[0050] The method specifically comprises the following steps:

[0051] S1 generates a unique hash value at each key link of Chinese medicinal materials such as planting, picking, processing, and warehousing; S2 records the hash value and corresponding metadata (such as time, location, operator, etc.) on the blockchain; S3 queries the hash value and its corresponding metadata on the chain when it is necessary to trace the quality information of Chinese medicinal materials; S4 builds a full life cycle traceability chain of Chinese medicinal materials based on metadata and displays it to consumers or managers.

[0052] Inventory optimization management algorithm: an inventory forecasting model combining moving average (MA) and seasonal adjustment factor (S). The forecasting model is defined as Q(t) = MA(t)*S(t), where MA(t) is the moving average of the tth period and S(t) is the seasonal adjustment factor of the tth period.

[0053] The method specifically comprises the following steps:

[0054] S1 First, collect past sales data from the historical sales database, including key indicators such as sales volume and return volume for each month or quarter. These data will serve as the basis for algorithm training and prediction; S2 Moving average calculation: Use the moving average method (MA) to smooth historical sales data to reduce the impact of seasonal fluctuations and other random factors on the prediction results. Set a fixed time window (such as sales volume in the last three or six months), calculate the average sales volume in the window, and obtain the moving average MA(t); S3 Determination of seasonal adjustment factor: Determine the seasonal adjustment factor S(t) based on the seasonal variation pattern in historical sales data. This usually requires periodic analysis of historical data to identify the impact of different seasons (such as spring, summer, autumn, and winter) on sales, and calculate the adjustment factor for each season accordingly; S4 Inventory forecasting: Multiply the moving average MA(t) by the seasonal adjustment factor S(t) to obtain the inventory demand Q(t) for the forecast period. The forecast value will be used as the basis for formulating inventory strategies and replenishment plans; S5 Dynamic Adjustment: In actual applications, the system will update the moving average and seasonal adjustment factor in real time based on the latest sales data and inventory status to achieve dynamic adjustment of inventory forecasts. In addition, the forecast model can be manually intervened according to special events (such as promotions, epidemic outbreaks, etc.) to improve the accuracy of the forecast.

[0055] Working principle: When a new batch of Chinese medicinal materials enters the warehouse, the RFID reader in the IoT sensing module reads the RFID tag information of the Chinese medicinal materials, and the temperature and humidity sensors, light sensors, etc. collect warehouse environmental data. These data are transmitted to the AI ​​processing module in real time. The intelligent storage location recommendation algorithm in the AI ​​processing module recommends a suitable storage location for the Chinese medicinal materials based on the characteristics of the Chinese medicinal materials and the current storage location of the warehouse. At the same time, the Chinese medicinal materials quality traceability model begins to record the storage start information of this batch of Chinese medicinal materials. During the storage process, the IoT sensing module continuously monitors environmental parameters, and the AI ​​processing module adjusts the warehouse's environmental control equipment (such as air conditioning, ventilation equipment, etc.) based on these data to ensure that the environment meets the storage requirements of Chinese medicinal materials. The inventory optimization management algorithm promptly reminds warehouse management personnel to replenish or adjust inventory based on sales data and inventory conditions. When Chinese medicinal materials are shipped out of the warehouse, the IoT sensing module records the shipping information, the AI ​​processing module updates the inventory data, and uploads the relevant information to the blockchain storage module. Consumers or regulatory authorities can query the full life cycle data of Chinese medicinal materials through the user interaction module, and the blockchain storage module ensures the authenticity and immutability of the data.

[0056] Embodiment 2

[0057] For the intelligent storage system for Chinese herbal medicines of the present invention, this embodiment further provides a mathematical optimization model:

[0058] (I) Application of Particle Swarm Optimization Algorithm in Warehouse Layout

[0059] Mathematical model and parameter formula definition

[0060] 1. Particle Representation

[0061] Particle position vector: xi = (xi1, xi2, ..., xid), where x ij Indicates the position of the i-th particle in the j-th dimension (representing the storage location or layout parameters of a certain type of cargo).

[0062] Particle velocity vector: vi = (vi1, vi2, ..., vid), represents the moving speed of the i-th particle in the search space.

[0063] 2. Fitness Function

[0064] The fitness function is used to evaluate the quality of particles, and its expression is:

[0065] f(x i )=α·SpaceUtilization(x i )+β·AccessEfficiency(x i )+γ·CostReduction(x i )

[0066] Among them, α, β, and γ are weight coefficients, which respectively represent the importance of space utilization, access efficiency, and cost reduction; SpaceUtilization(x i )、AccessEfficiency(x i )、CostReduction(x i ) are respectively scheme x i Quantified values ​​of space utilization, access efficiency and cost reduction.

[0067] 3. Parameter definition

[0068] Inertia weight (w): controls the influence of the particle's previous velocity on its current velocity, usually between [0, 1]. Self-cognition coefficient (c 1 ): The weight of the acceleration term of the particle moving to its own historical optimal position, which is used to adjust the influence of the individual optimal position on the particle speed. Social cognition coefficient (c 2 ): The weight of the acceleration term for particles to move toward the optimal position of the group, which is used to adjust the influence of the global optimal position on the particle speed.

[0069] 4. Speed ​​and position update formula

[0070] Update speed:

[0071] v ij (t+1)=w·v ij (t)+c 1 ·r 1 ·(p ij -x ij (t))+c 2 ·r 2 ·(g j -x ij (t))

[0072] Among them, r 1 and r 2 is a random number in the interval [0, 1], p ij is the jth dimension of the individual optimal position of the ith particle, g j is the jth dimension of the global optimal position. Update position:

[0073] x ij (t+1)=x ij (t)+v ij (t+1)

[0074] (II) Application of taboo search algorithm in inventory strategy Mathematical model and parameter formula definition

[0075] 1. Inventory policy representation

[0076] Inventory strategy vector: s = (s 1 ,s 2 , ..., s n ), where s i Represents some inventory parameter (such as order quantity, safety stock level, etc.).

[0077] 2. Objective Function

[0078] The objective function is used to measure the quality of the inventory strategy, and its expression is:

[0079] MinimizeZ=InventoryCost(s)+λ·StockoutRisk(s)

[0080] Among them, InventoryCost(s) represents the inventory cost, StockoutRisk(s) represents the out-of-stock risk, and λ is the weight coefficient of the out-of-stock risk.

[0081] 3. Taboo table and parameter definition

[0082] Taboo table: used to record the inventory strategies that have been visited to prevent the algorithm from falling into local optimality.

[0083] Taboo length (L): The number of iterations during which an element in the taboo table remains taboo.

[0084] Candidate set size (N): The number of candidate inventory policies generated in each iteration.

[0085] Through the above mathematical model and parameter formula definition, the application of the particle swarm optimization algorithm and the taboo search algorithm in the Chinese herbal medicine intelligent storage system of the present invention can be more clearly understood. These algorithms can effectively optimize the storage layout and formulate inventory strategies through the construction of mathematical models and the optimization of parameters, thereby improving the operational efficiency of the warehouse and the accuracy of inventory management.

[0086] The specific architecture of the intelligent storage system for Chinese herbal medicine based on particle swarm optimization and taboo search is as follows:

[0087] (I) Environment and logistics monitoring module

[0088] Temperature sensors, humidity sensors, light sensors, etc. are evenly distributed in the warehouse to monitor environmental parameters in real time. RFID reading and writing devices are installed at the warehouse entrances and shelves to automatically collect logistics information of Chinese herbal medicines. These devices transmit data to the local data collection server via Bluetooth or Wi-Fi.

[0089] (II) Intelligent Decision-making Module

[0090] The intelligent decision-making module runs on a high-performance computing server and loads the particle swarm optimization algorithm and taboo search algorithm model. After receiving the data from the environment and logistics monitoring module, the server starts the algorithm to optimize the warehouse layout and formulate inventory strategies.

[0091] For warehouse layout optimization, the particle swarm optimization algorithm continuously adjusts the position and speed of particles according to constraints such as the physical structure of the warehouse and the characteristics of the goods, and searches for the optimal combination of cargo storage locations.

[0092] For inventory strategy formulation, the taboo search algorithm searches for feasible inventory strategy space based on factors such as demand forecast, replenishment cycle, and cost constraints to avoid falling into local optimal solutions.

[0093] The decision results are output to the data storage and traceability module and the user interaction and visualization module.

[0094] (III) Data storage and traceability module

[0095] A distributed database is used to store a large amount of monitoring data and decision-making results, and blockchain technology is used to ensure the security and immutability of the data. Data is organized and stored according to time series and business classification, which is convenient for quick query and tracing.

[0096] (IV) User Interaction and Visualization Module

[0097] It includes a web-based management platform and mobile applications. The management platform provides comprehensive data analysis and management functions, and the mobile application facilitates users to conduct real-time monitoring and operation on site. The interface design is simple and intuitive, and uses charts, maps and other visual methods to display warehouse information.

[0098] Working principle: The environment and logistics monitoring module collects data in real time, the intelligent decision-making module performs optimization calculations, the results are stored in the data storage and traceability module, and users operate and monitor through the user interaction and visualization module.

[0099] Embodiment 3

[0100] For the Chinese herbal medicine intelligent storage system of the present invention, this embodiment further provides a mathematical optimization model

[0101] 1. Cargo space allocation optimization (simulated annealing algorithm)

[0102] Mathematical model:

[0103] Assume that the warehouse has N cargo spaces and M kinds of Chinese herbal medicines, and each Chinese herbal medicine has q i The number of units (i=1, 2, ..., M) that need to be stored. The goal of cargo space allocation is to maximize the storage efficiency of the warehouse while taking into account factors such as the convenience of picking up goods and the mutual influence between goods (such as moisture-proof and light-proof requirements).

[0104]

[0105] x ij It is a binary variable, indicating whether the i-th Chinese herbal medicine is stored in the storage location j (1 means stored, 0 means not stored).

[0106] Efficiency ij It is the efficiency coefficient of storing the i-th Chinese herbal medicine in shelf j, which is determined by factors such as the shelf location, size, and characteristics of the Chinese herbal medicine.

[0107] Cost ij is the cost coefficient for storing the i-th Chinese herbal medicine at storage location j, taking into account factors such as the difficulty of picking up goods and the mutual influence between goods.

[0108] Simulated annealing algorithm parameters:

[0109] Initial temperature T 0 : Set a higher initial temperature for extensive search when the algorithm starts.

[0110] Temperature drop function: such as T(k)=α·T(k-1), where α is a constant less than 1, and k is the number of iterations.

[0111] Inner loop times L k : At temperature Tk Next, execute L k Iterations of random perturbations and acceptance criteria.

[0112] Acceptance probability P: According to the Metropolis criterion, the probability of accepting a worse solution is

[0113] Where E new and E old are the objective function values ​​of the new solution and the old solution respectively.

[0114] 2. Inventory Control Optimization (Genetic Algorithm)

[0115] Mathematical model:

[0116] Inventory control problems usually involve decisions such as replenishment strategy, safety stock level, order quantity, etc. The goal here is to minimize inventory costs (including holding costs, ordering costs, and out-of-stock costs).

[0117]

[0118] C h , C o , C p They are the holding cost, ordering cost and stock-out cost per unit of Chinese herbal medicine.

[0119] Q i is the order quantity of the i-th Chinese herbal medicine.

[0120] s i It is the safety stock of the i-th Chinese herbal medicine.

[0121] h i is the unit holding cost rate of the i-th Chinese herbal medicine.

[0122] D i is the annual demand for the ith Chinese herbal medicine.

[0123] (D i -Q i ) + Indicates D i -Q i The positive part of , which is the out-of-stock quantity.

[0124] Genetic algorithm parameters:

[0125] Population size P: the number of individuals in the population in each generation.

[0126] Coding method: Binary coding, real number coding, etc. can be used. Select the appropriate method according to the type of decision variable.

[0127] Fitness function: directly use the opposite of the objective function as the fitness, that is

[0128] Fitness=-C.

[0129] Selection strategy: such as roulette selection, tournament selection, etc.

[0130] Crossover operator: single-point crossover, multi-point crossover, uniform crossover, etc. Select the appropriate crossover method according to the encoding method.

[0131] Mutation operator: randomly select a certain number of gene bits to flip.

[0132] Iteration number G: The total number of generations the algorithm runs.

[0133] Stop criteria: In addition to reaching the number of iterations, you can also set a fitness threshold or no significant improvement in fitness for multiple generations as a stop condition. Through these detailed mathematical models and parameter definitions, the Chinese herbal medicine intelligent warehousing system can more accurately optimize cargo space allocation and inventory control.

[0134] The specific architecture of the Chinese herbal medicine intelligent storage system based on simulated annealing and genetic algorithm is as follows:

[0135] (I) Data acquisition module

[0136] High-precision temperature and humidity sensors, light sensors, gas sensors, etc. are deployed in the warehouse, and RFID tags are attached to the packaging of Chinese herbal medicines. The sensors transmit data to the local data acquisition terminal through the Zigbee network, and the RFID tags communicate with the terminal through the reader / writer. After the terminal performs preliminary processing on the data, it sends it to the intelligent optimization module through Ethernet.

[0137] (II) Intelligent Optimization Module

[0138] The intelligent optimization module is deployed on the server and runs the simulated annealing algorithm and genetic algorithm program. After receiving the data from the data acquisition module, the server starts the algorithm to perform optimization calculations for cargo space allocation and inventory control.

[0139] For cargo space allocation optimization, the simulated annealing algorithm first generates an initial cargo space plan based on the warehouse layout and the characteristics of the goods, and then gradually finds the optimal cargo space allocation plan through continuous random perturbations and temperature control.

[0140] For inventory control optimization, the genetic algorithm first encodes the inventory control strategy and generates the initial population. Then, through fitness calculation, selection, crossover and mutation operations, the population is continuously evolved to finally find the optimal inventory control strategy.

[0141] The optimization results are sent to the blockchain storage module and the user interaction module.

[0142] (III) Blockchain storage module

[0143] The blockchain storage module consists of multiple nodes and uses distributed ledger technology to store data. Each node saves a complete copy of the data, and the consensus mechanism ensures the consistency and immutability of the data.

[0144] (IV) User Interaction Module

[0145] The user interaction module includes a web interface and a mobile application. The web interface provides comprehensive warehouse information display and management functions, suitable for office use. The mobile application facilitates warehouse managers to operate and query on site.

[0146] Working principle: The data acquisition module collects data in real time and transmits it to the intelligent optimization module. The intelligent optimization module uses algorithms to perform optimization calculations. The results are stored in the blockchain storage module. Users query and manage operations through the user interaction module.

[0147] The embodiments of the present invention are given for the purpose of illustration and description, and are not intended to be exhaustive or to limit the invention to the disclosed forms. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are selected and described in order to better illustrate the principles and practical applications of the present invention and to enable those of ordinary skill in the art to understand the present invention and thereby design various embodiments with various modifications suitable for specific uses.

Claims

1. An intelligent storage system for Chinese herbal medicine based on AI and blockchain, comprising an Internet of Things perception module, an AI processing module, a blockchain storage module and a user interaction module, wherein the Internet of Things perception module is used for data collection, the AI ​​processing module is used for data analysis and decision-making, the blockchain storage module is used to ensure the security and non-tamperability of data, and the user interaction module provides a friendly operation interface, characterized in that: The system can automatically optimize warehouse layout, achieve full quality traceability and dynamically adjust inventory levels.

2. The intelligent storage system for Chinese herbal medicine according to claim 1 is characterized in that: The AI ​​processing module includes an intelligent storage location recommendation algorithm, which uses a greedy algorithm combined with a graph neural network (GNN) to optimize the storage location.

3. The intelligent storage system for Chinese herbal medicine according to claim 1 is characterized in that: The blockchain storage module uses a hash function to construct a distributed ledger for storing unique hash values ​​generated in each link from planting, picking, processing to warehousing.

4. The intelligent storage system for Chinese herbal medicine according to claim 1, characterized in that: It also includes an inventory optimization management algorithm that combines a moving average (MA) with a seasonal adjustment factor (S) to dynamically adjust inventory levels based on historical sales data and seasonal changes.

5. The intelligent storage system for Chinese herbal medicine according to claim 1 is characterized in that: The system also includes AI image recognition and natural language processing technologies for effectively processing and analyzing unstructured data.

6. The intelligent storage system for Chinese herbal medicine according to claim 1, characterized in that: The Internet of Things perception module includes a sensor and an RFID tag.

7. The intelligent storage system for Chinese herbal medicine according to claim 1, characterized in that: The user interaction module provides an operation interface, allowing warehouse management personnel to view warehouse status, inventory levels, and cargo location distribution information, and make necessary adjustments and management.

8. The intelligent storage system for Chinese herbal medicine according to any one of claims 1 to 7, characterized in that: The system also includes automated and unmanned equipment.

9. The management method of the intelligent storage system of Chinese herbal medicine according to any one of claims 1 to 7, comprising: Collect real-time order data and inventory status information; Use graph neural network (GNN) to predict the visit frequency of each cargo location in the future; According to the cargo location selection function and the greedy strategy, the cargo location allocation is gradually optimized; the optimal cargo location allocation plan is output to guide the picking operation.

10. The management method of the intelligent storage system of Chinese herbal medicine according to any one of claims 1 to 7, comprising: Generate a unique hash value at each key link of Chinese medicinal materials planting, picking, processing, and storage; Record the hash value and corresponding metadata on the blockchain; When it is necessary to trace the quality information of Chinese medicinal materials, the full life cycle traceability chain of Chinese medicinal materials is constructed by querying the hash value and its corresponding metadata on the chain.

Images